Audio Source Separation is the process of separating a mixture (e.g. a pop band recording) into isolated sounds from individual sources (e.g. just the lead vocals).
|TREND||DATASET||BEST METHOD||PAPER TITLE||PAPER||CODE||COMPARE|
Models for audio source separation usually operate on the magnitude spectrum, which ignores phase information and makes separation performance dependant on hyper-parameters for the spectral front-end.
Ranked #16 on Music Source Separation on MUSDB18
This paper deals with the problem of audio source separation.
The thud of a bouncing ball, the onset of speech as lips open -- when visual and audio events occur together, it suggests that there might be a common, underlying event that produced both signals.
We study the use of the Wave-U-Net architecture for speech enhancement, a model introduced by Stoller et al for the separation of music vocals and accompaniment.
Ranked #9 on Speech Enhancement on DEMAND
We propose to add independent pseudo quantization noise to model parameters during training to approximate the effect of a quantization operator.
Ranked #13 on Language Modelling on WikiText-103
Given a multi-microphone recording of an unknown number of speakers talking concurrently, we simultaneously localize the sources and separate the individual speakers.
Based on this idea, we drive the separator towards outputs deemed as realistic by discriminator networks that are trained to tell apart real from separator samples.